A working project manager’s honest take on AI tools—what actually helps, what doesn’t, and why I’m not worried about being replaced.
From Curious to Daily User
I started experimenting with AI when ChatGPT launched in late 2022. Like most people, I was curious—asking it trivia questions, testing its limits, treating it more like a novelty than a tool. It was interesting, and I thought about how I could use it as a project manager.
As a PMP-certified project manager with 20+ years of experience, I’ve seen plenty of “revolutionary” tools come and go. Most created more work than they saved. So I approached AI the same way I approach any new tool: with healthy skepticism and a willingness to experiment. Over the past few years, I figured out where it actually helps—and where it doesn’t.
I now use AI daily. Not as a replacement for my judgment, but as a daily assistant. Here’s what I’ve learned.
What AI Actually Does Well for PMs
1. First Drafts of Standard Documents
AI excels at generating initial drafts of documents that follow predictable patterns:
- Project charters - Give it the project context and it produces a structured starting point
- Status report summaries - Feed it your bullet points and get polished prose
- Meeting agendas - Describe the meeting purpose and get a logical flow
- Communication templates - Stakeholder updates, kickoff announcements, milestone notifications
- Stakeholder registers - With enterprise Copilot, give it a list of names and it pulls email, department, and phone number from your organization’s directory
The key word is “first.” These drafts need review and refinement. But starting with 70% of the work done beats staring at a blank document.
2. Brainstorming and Expanding Thinking
When I’m stuck, AI helps me think through problems:
- Risk identification - “What risks should I consider for a software migration project in a regulated industry?” generates a comprehensive list I can filter and prioritize
- Stakeholder questions - “What questions might a CFO ask about this project proposal?” helps me prepare for the hard conversations
- Alternative approaches - “What are three different ways to structure this project?” gives me options I might not have considered
- Lessons learned - If your organization stores lessons learned in a system AI can access, it can surface relevant insights from past projects to help you avoid repeating mistakes or build on what worked
AI doesn’t replace my judgment about which risks matter most or which approach fits our organization. It expands the inputs I’m working with.
3. Translating and Simplifying
Technical projects require constant translation between audiences:
- Explaining technical concepts to executives
- Summarizing complex status for busy stakeholders
- Converting jargon-heavy vendor documentation into plain language
- Summarizing lengthy emails, email threads, or long chat conversations
AI handles this translation work quickly. I paste in a technical specification and ask for an executive summary. Catch up on a 50-message Teams thread in 30 seconds instead of scrolling through the whole thing. The output isn’t always perfect, but it’s a much faster starting point than reading or writing from scratch.
4. Real-Time Meeting Support
This one changed how I participate in meetings. Instead of frantically typing notes while trying to listen, I use Copilot during meetings to ask questions about topics as they come up. If someone mentions a technical concept I’m not familiar with, I can quickly get context without derailing the conversation or missing what comes next.
The result: I’m more present in meetings because I’m not splitting my attention between listening and note-taking. I capture what matters without missing the nuances that get lost when you’re heads-down typing.
After the meeting, I use AI to turn meeting transcripts into structured documentation—action items, decisions, open questions. This works especially well with Zoom meetings where transcripts are readily available. Feed the transcript to AI and ask for detailed meeting notes organized by topic. But the real win is staying engaged during the conversation itself.
What AI Does Poorly (Or Dangerously)
1. Organizational Context
AI doesn’t know that:
- The VP of Engineering and the VP of Product have a strained relationship
- Your organization’s “agile” is really waterfall with standups
- The last three projects with this vendor went poorly
- Budget approval actually requires the CFO’s EA to champion it
This context is everything in project management. AI produces generically correct advice that may be completely wrong for your specific situation.
2. Stakeholder Relationships
AI can help you prepare talking points, but it can’t:
- Read the room in a tense meeting
- Know when to push back and when to let something go
- Build the trust that gets projects unstuck
- Navigate the political dynamics that determine project success
The human side of project management—the part that actually determines outcomes—remains entirely human.
3. Judgment Calls
AI will give you options. It won’t tell you which option is right for your project, your organization, your constraints. That judgment comes from experience, and it’s the core value a PM provides.
4. Confidential Information
This is critical: Follow your company’s policy for what can and can’t be shared with AI.
Even if you have an enterprise AI tool, that doesn’t mean everything is fair game. Client names, financial data, employee information, strategic plans—your organization likely has policies about what can be shared and with which tools. Consumer AI tools are almost always off-limits for sensitive data, but even enterprise tools may have restrictions. Know your policy. When in doubt, ask.
My Daily AI Workflow
Here’s how AI fits into my actual work:
Morning planning (10 minutes):
- Get prepared for key meetings
- Draft any status communications I need to send
During work:
- Use AI to brainstorm when I’m stuck
- Generate first drafts of documents I need to create
- Translate technical content for different audiences
- Reword my email messages or sentences
- Turn rough notes or transcripts into structured meeting notes
- Format Excel files, create complex formulas, and set up conditional formatting
End of day (5 minutes):
- Process meeting notes into action items and summaries
- Find any action items assigned to me
- Draft follow-up emails
What I don’t use AI for:
- Anything with confidential data
- Final versions of important communications (I always rewrite in my voice)
- Decisions (AI informs, I decide)
- Relationship-dependent work
The “Will AI Replace PMs?” Question
Short answer: No.
Longer answer: AI will replace PMs who only do documentation and administrative work. But that was never the valuable part of project management anyway.
The valuable PM work is:
- Judgment - Deciding what to do when the plan meets reality
- Relationships - Building trust that makes collaboration work
- Navigation - Moving projects through organizational complexity
- Synthesis - Making sense of ambiguous situations
AI makes the administrative parts faster, which means more time for the parts that actually matter. That’s not a threat—it’s an upgrade.
Getting Started
If you’re a PM who hasn’t tried AI tools yet, here’s my suggestion:
Week 1: Start with meeting notes. Feed a transcript or your rough notes into AI and ask for structured meeting notes with action items. This is low-risk and immediately useful—you’ll see the value right away.
Week 2: Before a difficult meeting, ask AI to help you anticipate questions and prepare responses.
Week 3: Use AI to draft a small document you’d normally write from scratch—a one-page status summary or a short stakeholder update. AI handles shorter documents well; for larger documents, you’ll need to work in sections.
Week 4: Experiment with brainstorming—ask AI to help you identify risks, stakeholder concerns, or alternative approaches.
Start small. See what helps. Ignore what doesn’t. Develop your own workflow based on what actually makes your work better.
The Bottom Line
AI is a tool. Like any tool, it’s useful for some things and useless for others. The PMs who figure out where it helps—and where it doesn’t—will be more effective than those who either ignore it entirely or try to use it for everything.
I use AI daily now. It makes me faster at the administrative parts of my job, which gives me more time for the parts that actually require a human project manager. That’s not a revolution—it’s just a better set of tools.
And I’m still the one doing the project management.
Related Posts
- Coming in Week 2: 5 PM Tasks AI Actually Does Well (And 5 It Doesn’t)
- Coming in Week 3: Your First Week Using AI as a Project Manager
AI won’t replace project managers. But project managers who use AI effectively will have an advantage over those who don’t. The goal isn’t to replace yourself—it’s to level up by offloading the parts that were never the valuable part anyway.